Causal discovery algorithms: A practical guide

2017 ◽  
Vol 13 (1) ◽  
pp. e12470 ◽  
Author(s):  
Daniel Malinsky ◽  
David Danks
Keyword(s):  
Causal Discovery ◽  
Practical Guide ◽  
Discovery Algorithms

scholarly journals Comment on "Estimating causal networks in biosphere–atmosphere interaction with the PCMCI approach"

2021 ◽  
Author(s):  
Jarmo Mäkelä ◽  
Laila Melkas ◽  
Ivan Mammarella ◽  
Tuomo Nieminen ◽  
Suyog Chandramouli ◽  
...  
Keyword(s):  
Domain Knowledge ◽  
Prior Information ◽  
Causal Model ◽  
Concept Drift ◽  
Causal Structure ◽  
Causal Discovery ◽  
Causal Networks ◽  
Atmosphere Interaction ◽  
Model Structures ◽  
Discovery Algorithms

Abstract. This is a comment on "Estimating causal networks in biosphere–atmosphere interaction with the PCMCI approach" by Krich et al., Biogeosciences, 17, 1033–1061, 2020, which gives a good introduction to causal discovery, but confines the scope by investigating the outcome of a single algorithm. In this comment, we argue that the outputs of causal discovery algorithms should not usually be considered as end results but starting points and hypothesis for further study. We illustrate how not only different algorithms, but also different initial states and prior information of possible causal model structures, affect the outcome. We demonstrate how to incorporate expert domain knowledge with causal structure discovery and how to detect and take into account overfitting and concept drift.

scholarly journals Integrating Overlapping Datasets Using Bivariate Causal Discovery

2020 ◽  
Vol 34 (04) ◽  
pp. 3781-3790
Author(s):  
Anish Dhir ◽  
Ciaran M. Lee
Keyword(s):  
Conditional Independence ◽  
Causal Structure ◽  
Real Data ◽  
Causal Discovery ◽  
Causal Relations ◽  
Multiple Datasets ◽  
Single Dataset ◽  
Markov Equivalence ◽  
Causal Structures ◽  
Discovery Algorithms

Causal knowledge is vital for effective reasoning in science, as causal relations, unlike correlations, allow one to reason about the outcomes of interventions. Algorithms that can discover causal relations from observational data are based on the assumption that all variables have been jointly measured in a single dataset. In many cases this assumption fails. Previous approaches to overcoming this shortcoming devised algorithms that returned all joint causal structures consistent with the conditional independence information contained in each individual dataset. But, as conditional independence tests only determine causal structure up to Markov equivalence, the number of consistent joint structures returned by these approaches can be quite large. The last decade has seen the development of elegant algorithms for discovering causal relations beyond conditional independence, which can distinguish among Markov equivalent structures. In this work we adapt and extend these so-called bivariate causal discovery algorithms to the problem of learning consistent causal structures from multiple datasets with overlapping variables belonging to the same generating process, providing a sound and complete algorithm that outperforms previous approaches on synthetic and real data.

scholarly journals A Ladder of Causal Distances

2021 ◽  
Author(s):  
Maxime Peyrard ◽  
Robert West
Keyword(s):  
Real World ◽  
Causal Model ◽  
Ground Truth ◽  
Causal Models ◽  
Causal Discovery ◽  
Major Significance ◽  
Benchmark Datasets ◽  
Real World Datasets ◽  
Discovery Algorithms ◽  
Graphical Structure

Causal discovery, the task of automatically constructing a causal model from data, is of major significance across the sciences. Evaluating the performance of causal discovery algorithms should ideally involve comparing the inferred models to ground-truth models available for benchmark datasets, which in turn requires a notion of distance between causal models. While such distances have been proposed previously, they are limited by focusing on graphical properties of the causal models being compared. Here, we overcome this limitation by defining distances derived from the causal distributions induced by the models, rather than exclusively from their graphical structure. Pearl and Mackenzie [2018] have arranged the properties of causal models in a hierarchy called the ``ladder of causation'' spanning three rungs: observational, interventional, and counterfactual. Following this organization, we introduce a hierarchy of three distances, one for each rung of the ladder. Our definitions are intuitively appealing as well as efficient to compute approximately. We put our causal distances to use by benchmarking standard causal discovery systems on both synthetic and real-world datasets for which ground-truth causal models are available.

Sign in / Sign up

Export Citation Format

Share Document